🔥 Inspiration
The inspiration came from a problem we all face: people move to new cities for study, work, or travel… and then discover unexpected surprises like noise, insects, poor water quality, or unsafe neighborhoods. There’s no platform that gives a real, honest, holistic picture of life quality in any neighborhood — you might know the weather, but not the pollution. You might see beautiful photos, but not crime waves. You might love a place online, then regret living there.
So we built CitySense to solve that.
CitySense analyzes any city or neighborhood and generates a detailed AI-powered urban quality report, showing everything that actually matters before choosing where to live.
🚀 What CitySense provides
CitySense doesn’t stop at basic city data — it reveals what people really need to know:
Noise level variations day vs night
Air & water quality indicators
Mosquito and insect density patterns
Traffic congestion + roadwork alerts
Neighborhood safety & crime probability
Average rent prices & cost of living
Weather behavior + seasonal patterns
Verified user reviews & local insights
A visual AI heatmap showing livability score
With one search — you know the full picture.
đź› How we built it
We built CitySense using a scalable, AI-first architecture designed to handle data in motion:
Component Technology
Core Framework Django Frontend HTML, CSS, JavaScript Weather & environmental data Open-Meteo API + scraped datasets Real-time streams Confluent Kafka (for heatmap processing) AI layer LLM summarization + anomaly detection
The system collects raw data → filters it → detects anomalies → generates a clean AI summary optimized for human decisions.
⛓ Challenges we faced
Urban analytics is complex. Our main challenges were:
Collecting reliable data for cities with limited public datasets
Integrating multiple APIs smoothly under time pressure
Building a real-time heatmap using Confluent streams
Making AI summaries consistent, verifiable & bias-free
Designing a simple UI for a massive data model
Ensuring fast performance despite large geographic layers
Balancing visual design vs analytical depth
But we pushed through each one — fast.
🏆 Achievements we’re proud of
Built a fully operational MVP in a very short time
One of the first address-level city analyzers in the region
Designed a clean and intuitive UX for a complex concept
Successfully integrated AI with live streaming data
Created a visual heatmap that makes city risks instantly noticeable
Turned raw messy data → into clarity that anyone can understand
We didn’t just build a tool — we built a decision-making engine.
đź“š What we learned
Real-time analytics at city scale is harder than it looks
AI is powerful — but requires careful prompting and validation
Users value trust & credibility more than fancy UI
GIS data must be simplified for speed and usability
Good data visualization can replace 20 paragraphs of text
Insight is more valuable than raw numbers
🔮 What’s next
We’re just getting started. Next step is expanding CitySense into a global intelligent atlas:
Support for more cities and districts worldwide
Richer layers: weather, safety, pollution, water health, services
AI-powered housing recommendations
Predictive climate-risk modeling
Launching a dedicated mobile app
Personalized user dashboards + favorites tracking
Bilingual interface for international adoption
CitySense isn’t a website — it’s the future of how humans choose where to live.
Log in or sign up for Devpost to join the conversation.